Until recently, gene therapy protocols have often relied on vectors derived from retroviruses, such as murine leukemia virus (MLV). These vectors are useful because the genes they transduce are integrated into the genome of the target cells, a desirable feature for long-term expression. However, these retroviral vectors can only transduce dividing cells, which limits their use for in vivo gene transfer in nonproliferating cells, such as hepatocytes, myofibers, hematopoietic stem cells, and neurons.
Lentiviruses are a type of retrovirus that can infect both dividing and nondividing cells. They have proven extremely efficient at providing long-term gene expression (for up to 6 months) in a variety of nondividing cells (such as, neurons and macrophages) in animal models. See, e.g., Amado et al., Science 285:674-676 (July 1999). It has been proposed that the optimal gene transfer system would include a vector based on HIV, or other lentivirus, that can integrate into the genome of nonproliferating cells. Because retroviruses integrate in the genome of the target cells, repeated transduction is unnecessary. Therefore, in contrast to an adenoviral vector capable of in vivo gene delivery, problems linked to the humoral response to injected viral antigens can be avoided. See, e.g., Naldini et al., Science, 272:263-267 (1996), p. 263.
HIV and other lentiviruses have a complex genome that, in addition to the essential structural genes (env, gag, and pol), contains regulatory (tat and rev) and accessory genes (vpr, vif vpu, and nef). HIV has evolved to efficiently infect and express its genes in human cells, and is able to infect nondividing cells such as macrophages because its preintegration complex can traverse the intact membrane of the nucleus in the target cell. This complex contains, in addition to the viral DNA, the enzyme integrase, the product of the vpr gene, and a protein encoded by the gag gene called matrix. The matrix protein enables the preintegration complex to pass into the nucleus to access the host DNA. Lentiviruses cannot efficiently transduce truly quiescent cells (cells in the G0 state). However, unlike murine retroviral vectors, in addition to being able to infect dividing cells, HIV-based vectors can achieve effective and sustained transduction and expression of therapeutic genes in nondividing cells, such as hematopoietic stem cells and in terminally differentiated cells such as neurons, retinal photoreceptors, muscle, and liver cells. See, e.g., Amado et al. (July 1999) and Klimatcheva et al., Frontiers in Bioscience 4:d481-496 (June 1999), and the references cited therein.
Although lentiviral vectors can be efficient gene delivery vehicles, there are safety concerns due to their origin. Therefore, the field has turned its attention to the development of vectors and production systems with built-in safety features to prevent the emergence of replication competent lentivirus (RCL). For example, in most laboratory applications, lentiviral vectors are generally created in a transient system in which a cell line is transfected with three separate constructs: a packaging construct, a transfer construct, and an envelope encoding construct. The packaging construct contains the elements necessary for vector packaging (except for env) and the enzymes required to generate vector particles. The transfer construct contains genetic cis-acting sequences necessary for the vector to infect the target cell and for transfer of the therapeutic (or reporter) gene. The lentivirus env gene is generally deleted from the packaging construct and instead the envelope gene of a different virus is supplied in a third vector “the env-coding vector”, although the lentiviruses env gene may be used if it is desired that the vector be intended to infect CD4+T cells. A commonly used envelope gene is that encoding the G glycoprotein of the vesicular stomatitis virus (VSV-G), which can infect a wide variey of cells and in addition confers stability to the particle and permits the vector to be concentrated to high titers (see, e.g., Naldini et al., Science 272:263-267 (1996) and Akkina et al. J. Virol. 70:2581 (1996). The use of three separate constructs and the absence of overlapping sequences between them minimizes the possibility of recombination during lentivirus (transfer) vector production. In addition, because no viral proteins are expressed by the lentiviral (transfer) vector itself, they do not trigger an effective immune response against cells expressing vector in animal models (a particular problem with vectors based on adenovirus). See, e.g., Amado et al., Science 285:674-676 (July 1999) and the references cited therein. See also Naldini et al. Science 272:263-267 (1996).
The initial packaging plasmids contained most HIV genes except for env. In an effort to improve safety, subsequent HIV vectors have been produced in which the packaging plasmid is devoid of all accessory genes. This process does not interfere with efficient vector production and significantly increases the safety of the system because potential RCLs lack the accessory genes necessary for efficient replication of HIV in humans. Although these vectors can transduce growth-arrested cell lines and neurons in vivo, they have been reported to not efficiently transduce macrophages. The accessory gene vpr is believed to be necessary for HIV infection of these cells using these HIV vectors. See, Zufferey et al., Nature Biotechnol. 15:871-875 (1997). In contrast, as discussed later herein, the HIV-based lentiviral vectors of the present invention do not need any HIV accessory genes in order to be able to infect human macrophages and the other cells tested.
The requirement of vpr or vif for efficient transduction of liver cells has also been reported. See, e.g., Kafri et al., Nature Genet. 17:314 (1997). These results indicate that the requirement of accessory genes for efficient lentivirus-mediated gene transfer is dependent on the type of cell chosen as target, suggesting that future applications of lentiviral vectors may involve vector constructs with different accessory genes, as needed.
Zufferey et al., (1997) describe an HIV vector system in which the virulence genes, env, vif vpr, vpu, and nef have been deleted. This multiply attenuated vector conserved the ability to transduce growth-arrested cells and monocyte-derived macrophages in culture, and could efficiently deliver genes in vivo into adult neurons. The packaging plasmids described Zufferey et al. (1997) and Naldini et al. (1996) encode Rev and Tat, in addition to Gag and Pol.
Lentiviral vectors engineered to become packaged into virions in the absence of the regulatory gene tat have also been described. See, e.g., Kim et al., J. Virol. 72:811-816 (1998) and Miyoshi et al. J. Virol. 72:8150-8157 (1998). In these vectors the tat gene has been removed from the packaging plasmid. Kim et al. state that tat is not necessary as long as the serial 5′ LTR promoter is replaced with a strong constitutive promoter. It also has other advantages for HIV therapy. Replacement of the HIV-1 LTR with a constitutive HCMV promoter permits the use of anti-Tat molecules such as Tat transdominant mutants or Tat activation response element decoys as therapeutic agents, since they will not affect vector production. (see p. 814, col. 2). The removal of the tat gene eliminates an essential virulence factor that could contribute to a possible RCL. Kim et al. (1998) describe a vector system which does not contain tat, vif, vpr, vpu and nef. The preferred vector system includes the rev gene which, the authors state “with RRE, is required for efficient RNA handling in this system.” (p. 811, col. 2). However, Kim et al. also constructed Rev independent constructs using CTE. Kim et al. state that the rev/RRE components could be removed by using a sequence such as the Mason-Pfizer monkey virus (MPMV) constitutive transport element (CTE), thereby eliminating all accessory proteins, but this leads to a significant reduction in titer.
Srinivasakumar et al., J. Virol. 71:5841-5848 (1997) describes the generation of stable HIV-1 packaging lines that constitutively express high levels of HIV-1 structural proteins in either a Rev-dependent or a Rev-independent fashion. These cell lines were used to assess gene transfer by using a HIV-1 vector expressing the hygromycin B resistance gene and to study the effects of Rev, Tat, and Nef on the vector titer. The Rev-independent cell lines were created by using gag-pol and env expression vectors that contain the MPMV CTE. This article describes the construction of four plasmids, among others: CMV gagpol-RRE and pCMVenv, which require Rev coexpression for HIV-1 structural gene expression, and pCMV gagpol-CTE and pCMVenv-CTE, which do not. To create Rev-containing and Rev-independent packaging, cell lines, CMT3 cells were transfected with vectors expressing Gag, Gag-Pol, and Env, using a calcium phosphate transfection procedure.
By creating an HIV vector which contained the MPMV CTE (pTR167-CTE) and a packaging cell line which expressed the HIV structural proteins in a Rev-independent fashion, the authors were able to obtain a HIV vector system that functions completely without Rev. The titer of the vector obtained from this system was essentially the same as that obtained from a parallel system which contained Rev. The authors state that, in this context, the CTE seemed to substitute completely for Rev-RRE functions, similar to what was previously observed in transient-expression assays with Rev-dependent constructs. This is in contrast to situations where several rounds of HIV replication were measured. In those cases, titers from CTE-containing viruses were always reduced by at least 1 log unit compared to viruses utilizing Rev and the RRE. (See, Srinivasakumar et al., p. 5847).
The authors state that the advantages of having a HIV vector system that works in the absence of Rev opens the possibility of using it as a delivery vehicle for intracellular immunization against Rev function. Genes encoding Rev antagonists that have dramatic inhibitory effects on HIV replication, such as Rev M10 or RRE decoys, could be introduced into an HIV vector and put into cells normally injectable by HIV. Expression of the “anti-Rev” gene would be expected to dampen HIV infection. Any residual HIV replication should lead to activation of the vector LTR (by Tat) and create a vector-derived RNA that would be packaged by proteins derived from the infectious virus. In this scenario, the wild-type virus would act as a helper that may allow the spread of vector particles to previously nonimmunized cells. Because of the additional vector spread, it is likely that this type of scheme will be more effective in modulating HIV infection in vivo than one based on traditional retrovirus vectors. The authors state that they are currently testing this approach in model systems. (See, Srinivasakumar et al., p. 5847).
Another development in the quest for a safe system is the so-called self-inactivating (SIN) vector. See, e.g., Yu et al., Proc Natl Acad Sci USA 83:3194-8 (1986) and Miyoshi et al., J. Virol. 72:8150 (1998). In Yu et al., a retrovirus-derived vector SIN vector was designed for the transduction of whole genes into mammalian cells. The SIN vector of Yu et al. contains a deletion of 299 base pairs in the 3′ long terminal repeat (LTR), which includes sequences encoding the enhancer and promoter functions. When viruses derived from such vectors were used to infect NIH 3T3 cells, the deletion was transferred to the 5′ LTR, resulting in the transcriptional inactivation of the provirus in the infected cell. Introduction of a hybrid gene (human metallothionein-promoted c-fos) into cells via a SIN vector was not associated with rearrangements and led to the formation of an authentic mRNA transcript, which in some cases was induced by cadmium. The vector described in Miyoshi et al. also contains a deletion the 3′ (downstream) LTR. A sequence within the upstream LTR serves as a promoter under which the viral genome is expressed. The deletion introduced in the downstream LTR is transferred to the upstream LTR during reverse transcription. This deletion inactivates the LTR promoter and eliminates the production of vector RNA. The gene (or genes) to be transferred (e.g., a reporter or therapeutic gene) is expressed from an exogenous viral or cellular promoter that is inserted into the lentivirus vector. An important safety feature of SIN vectors is that inactivation of the promoter activity of the LTR reduces the possibility of insertional mutagenesis (of the transfer vector) into the host genome. In addition, because the expression of the (transfer) vector RNA is eliminated, the potential for RCL production in the target cell is further minimized. SIN vectors should be particularly useful in gene transfer experiments designed to study the regulated expression of genes in mammalian cells. Absence of enhancer and promoter sequences in both LTRs of the integrated provirus should also minimize the possibility of activating cellular oncogenes and may provide a safer alternative to be used in human gene therapy. Other modifications to enhance safety and specificity include the use of specific internal promoters that regulate gene expression, either temporally or with tissue or cell specificity.
Other strategies to improve safety in human studies would be to use nonhuman lentiviruses such as simian immunodeficiency virus, bovine immunodeficiency virus, or equine infectious anemia virus. Of these, vectors derived from the feline immunodeficiency virus have been engineered to efficiently transduce nondividing human cells. See, e.g., Poeschla et al., Nature Med. 4:354-357 (1998) and WO 99/15641. In addition, White et al., J. Virol. 73:2832-2840 (April 1999) described lentiviral vectors using human and simian immunodeficient virus elements in attempt to improve safety by reducing the likelihood of recombination between packaging constructs and transfer constructs.
The development of efficient packaging lines has proven challenging because expression of the VSV-G envelope and a number of HIV proteins is toxic to cells. Recently, a producer line has been designed in which the expression of packaging genes and VSV-G, and therefore the production of vector, can be turned on at will. Kafri et al., J. Virol. 73-576-584 (1999). The cell line can be expanded for scale-up vector production when the expression of toxic genes is turned off. This cell line produces high titer vector without generating RCL. Hematopoietic stem cells transduced with an HIV vector were transplanted into rhesus macaques as described by Donahue et al. Blood 92 (suppl. 1), abstract 4648.5 (1998) with at least a 14-month follow-up. At that time the procedure proved to be safe; all animals in the study have remained healthy without evidence of circulating HIV or vector. See, Amado et al., Science 285:674-676 (July 1999).
Many gene therapy protocols have been designed to correct a number of inherited metabolic, infectious, or malignant diseases using the hematopoietic stem cell. This cell has the capacity to self-renew and to differentiate into all of the mature cells of the blood and immune systems. Many diseases that affect these systems could potentially be treated by the stable introduction of therapeutic genes into stem cells. Recently, lentiviral vectors were shown to bypass the need for ex vivo stem cell stimulation (which is necessary when using murine retroviral vectors), by mediating efficient gene transfer into very primitive human stem cells that contributed to stable, long-term reconstitution of SCID mouse bone marrow with many hematopoietic lineages. See, e.g., Miyoshi et al., Science 283:682 (1999). Similarly, in a rhesus macaque model of autologous transplantation with lentivirus-transduced stem cells, multilineage gene expression was found, suggesting transduction of an early blood cell progenitor under conditions of minimal stem cell stimulation, ordinarily insufficient for transduction with murine retroviruses. See, Donahue et al., Blood 92 (suppl. 1), abstract 4648.5 (1999) and Amado et al., Science 285:674-676 (July 1999).
In HIV infection, another advantage of lentiviral vectors designed against HIV is their potential to be mobilized by HIV in the infected patient, because the virus supplies all of the necessary elements for packaging of the vector. If these mobilized vectors contained the HIV envelope, they could efficiently transfer their genes (for example, genes custom-designed to confer resistance against HIV) into CD4+ T cells, protecting them from subsequent HIV infection. Lentiviral vectors can also be designed to efficiently express their genes only in CD4+ T cells that are infected with HIV (so called tat-inducible vectors). In these vectors, all HIV genes, including tat and rev, are ablated; cis-acting sequences required for integration, expression, and packaging are retained, and expression is dependent on the activity of the HIV LTR (which requires transactivation by Tat). It has been shown that in this system, vector expression is induced efficiently upon HIV infection. Moreover, in the absence of genes that confer resistance against HIV, stable integration of this vector in permissive cell lines resulted in inhibition of HIV replication. Although the mechanism of HIV inhibition has not been completely elucidated, preliminary results suggest that this vector competes with HIV at the level of reverse transcription. See, An et al., J. Virol., in press, and Amado et al., Science 285:674-676 (1999).
A number of other potential medical applications, where the modification of the genetic material of quiescent cells could result in the prevention or reversal of a disease process, are beginning to be explored. For example, the finding that lentiviral vectors can mediate stable and long-term gene transfer by direct injection of vector into the rat and mouse retina has lent support to the notion of gene therapy for the treatment of retinitis pigmentosa. This degenerative disease of the retina is characterized by photoreceptor cell death, resulting in a slow progression to blindness. Mutations in the cGMP phosphodiesterase β subunit (PDEβ) gene of rod photoreceptors lead to an autosomal recessive form of retinitis pigmentosa in humans, and in the rd mouse model of the disease. Previous studies have shown that adenovirus and adeno-associated virus-mediated PDEP subretinal gene transfer results in a delay in photoreceptor cell death. Using the rd mouse model, a recent study demonstrated that photoreceptors could be rescued in up to 50% of eyes injected with a lentivirus vector containing the murine PDEβ gene. In contrast with the short-term expression previously obtained with adenovirus vectors, PDEβ expression in this study persisted for at least 24 weeks. This finding points to the potential success of gene therapy in a disease that currently lacks effective treatment. See, Takahashi et al., J. Virol., 73:7812-7816 (September 1999) and Amado et al. Science, 285:674-676 (1999).
In nature, the expression of gag, pol, and env of HIV-1 depends on the presence of the viral Rev protein. This dependence is, at least in part, due to the presence of negatively acting sequences (inhibitory or instability elements [INS]) located within unspliced and partially spliced mRNAs. The positive interaction of Rev with the Rev-responsive element [RRE] in these mRNAs counteracts the negative effects of the inhibitory sequences.
None of the above references teach or suggest that the gag and/or pol genes described therein may be replaced with the gag and/or pol genes in which the inhibitory/instability have been mutated to render their expression Rev-idependent. Furthermore, there is no disclosure of the specific HIV-1 gag/pol or SIV gag mutated genes described herein.
The gag/pol clone of the invention was made using the method for eliminating inhibitory/instability regions from a gene as first described in U.S. patent application Ser. No. 07/858,747, filed Mar. 27, 1992 (which issued as U.S. Pat. No. 6,174,666) entitled “Method of Eliminating Inhibitory/Instability Regions from mRNA” and later described in a Continuation-in-Part (“CIP”) application, filed as PCT application PCT/US93/02908 on Mar. 29, 1993 and U.S. Pat. Nos. 5,972,596 and 5,965,726. The disclosure of the CIP application was published as International Publication No. WO 93/20212 on Oct. 14, 1993. (The disclosures of these patents and patent applications are specifically incorporated by reference herein in their entirety.) The method was also described in Schwartz et al., J. Virol. 66:7176-7182 (1992).
Schneider et al., J. Virol. 71:48924903 (1997), extend the work described in the patent applications and in Schwartz et al. by identifying and characterizing additional INS within gag, protease and pol genes and mutating them in a similar manner. Schneider et al. disclose nucleic acid constructs which contain completely mutated HIV-1 gag genes, but only partially mutated HIV-1 pol genes.
Schneider et al. demonstrate that expression vectors containing an intact or nearly intact p55gag region allow the production of immature viral particles in mammalian cells in the absence of any other HIV proteins. The introduction of additional mutations in the protease region allowed efficient production of Gag/protease, which resulted in processing of the Pr55gag precursor and production of mature Gag particles with a lentivirus-like conical-core structure.
Schneider et al. disclose that Rev-independent expression vectors allow the efficient expression of Gag proteins in many cell lines that are not able to support efficient Rev-RRE-dependent rescue of these RNAs. Schneider et al. also disclose that gag/pol expression vectors may be important for vaccination approaches against HIV-1, since the gag/pol region is more conserved than is the env region and may be important for an effective immune response against HIV and for protection against infection. They also state that efficient HIV gene expression in many cells is also of interest for possible gene transfer experiments using lentiviral vectors in nondividing or slowly dividing cells, since HIV and the other lentiviruses are able to infect quiescent cells.
Pavlakis et al., Natl Conf Hum Retroviruses Relat Infect (2nd). (1995), 91, state that Rev-independent Gag expression vectors were able to produce viral particles in human and mouse cells in the absence of any other HIV proteins, and that additional mutations in the pol region allowed the expression of the protease and the processing of the p55 gag precursor. Direct DNA injection of TAT and Rev independent Gag expression vectors in mouse muscle resulted in Gag expression detected by ELISA and in anti-gag antibody response. Several Rev- and Tat-independent Gag expression cassettes were inserted into retroviral vectors and cell lines expressing Gag or Gag fragments that are dominant negative inhibitors of HIV-1 were constructed.
Shiver et al. (1996) describe the results of DNA vaccination of mice and non-human primates with mutated plasmid DNA encoding either mutated genes encoding HIV-1 gag (p55 gag) or env (gp120 or gp160). Both gag and env vaccine recipients exhibited antigen-specific cytotoxic and helper T lymphocyte (CTL, Th) responses. The results are stated to demonstrate that DNA vaccines elicited long-lived T cell responses in both mice and nonhuman primates that were disseminated throughout the lymphatics.